Planetary and Space Science 85 (2013) 299–311

Contents lists available at ScienceDirect

Planetary and Space Science

0032-06http://d

☆BaseINT 2.5Tautenbin Roma

n CorrCorreosTel.: +3

E-movidiuv

journal homepage: www.elsevier.com/locate/pss

739 observed NEAs and new 2–4 m survey statisticswithin the EURONEAR network$

O.Vaduvescu a,b,c,n, M. Birlan b, A. Tudorica d,e, M. Popescu b,f,h, F. Colas b, D.J. Asher j,A. Sonka g,h, O. Suciu i,k, D. Lacatus l,m,p, A. Paraschiv l,m,p, T. Badescu n, O. Tercu o,p,A. Dumitriu o,p, A. Chirila o,p, B. Stecklumq, J. Licandro c,r, A. Nedelcu b,f, E. Turcu s,F. Vachier b, L. Beauvalet b, F. Taris b, L. Bouquillon b, F. Pozo Nunez t, J.P. Colque Saavedra u,E. Unda-Sanzana u, M. Karami a,v,w, H.G. Khosroshahi v, R. Toma j,k, H. Ledo a,z, A. Tyndall x,a,L. Patrick a,y, D. Föhring a,aa, D. Muelheims n, G. Enzian n, D. Klaes n, D. Lenz n, P. Mahlberg n,Y. Ordenes n, K. Sendlinger n

a Isaac Newton Group of Telescopes (ING), Apartado de Correos 321, E-38700 Santa Cruz de la Palma, Canary Islands, Spainb IMCCE, Observatoire de Paris, 77 Avenue Denfert-Rochereau, 75014 Paris Cedex, Francec Instituto de Astrofísica de Canarias (IAC), C/Vía Láctea s/n, 38205 La Laguna, Spaind Bonn Cologne Graduate School of Physics and Astronomy, Germanye Argelander-Institut für Astronomie, Universität Bonn, Auf dem Hügel 71, D-53121 Bonn, Germanyf The Astronomical Institute of the Romanian Academy, Cutitul de Argint 5, 040557 Bucharest, Romaniag The Astronomical Observatory “Admiral Vasile Urseanu”, B-dul Lascar Catargiu 21, Bucharest, Romaniah Bucharest Astroclub, B-dul Lascar Catargiu 21, sect 1, Bucharest, Romaniai Babes-Bolyai University, Faculty of Mathematics and Informatics, 400084 Cluj-Napoca, Romaniaj Armagh Observatory, College Hill, Armagh BT61 9DG, Northern Ireland, United Kingdomk Romanian Society for Meteors and Astronomy (SARM), CP 14 OP 1, 130170 Targoviste, Romanial Research Center for Atomic Physics and Astrophysics, Faculty of Physics, University of Bucharest, Atomistilor 405, CP Mg-11, 077125 Magurele, Ilfov, Romaniam Institute of Geodynamics Sabba S. Stefanescu, Jean-Louis Calderon 19-21, Bucharest RO-020032, Romanian Rheinische-Friedrich-Wilhelms Universitaet Bonn, Argelander-Institut fur Astronomie, Auf dem Hugel 71, D-53121 Bonn, Germanyo Galati Astronomical Observatory of the Natural Sciences Museum Complex, Str. Regiment 11-Siret, no 6A, 800340 Galati, Romaniap Calin Popovici Astronomy Club, Str. Regiment 11-Siret, no 6A, 800340 Galati, Romaniaq Türinger Landessternwarte Tautenburg, Sternwarte 5, D-07778 Tautenburg, Germanyr Departamento de Astrofísica, Universidad de La Laguna, E-38205 La Laguna, Tenerife, Spains The Astronomical Observatory of “Stefan cel Mare” University of Suceava, Str. Universitatii 13, 720229 Suceava, Romaniat Instituto de Astronomia, Universidad Catolica del Norte, Avenida Angamos 0610, Antofagasta, Chileu Unidad de Astronomia, Universidad de Antofagasta, Avda. Angamos 601, 1270300 Antofagasta, Chilev School of Astronomy, Institute for Research in Fundamental Sciences (IPM), PO Box 19395-5531, Tehran, Iranw Sharif University of Technology, Azadi Ave., PO Box 11365-11155, Tehran, Iranx Jodrell Bank Centre for Astrophysics, School of Physics and Astronomy, University of Manchester, M13 9PL, United Kingdomy Department of Physics and Astronomy, University of Sheffield, Sheffield S3 7RH, United Kingdomz Centre for Astrophysics Research, University of Hertfordshire, College Lane, Hatfield AL10 9AB, United Kingdomaa Department of Physics, Centre for Advanced Instrumentation, University of Durham, South Road, Durham DH1 3LE, United Kingdom

a r t i c l e i n f o

Article history:Received 12 December 2012Received in revised form7 May 2013Accepted 27 June 2013Available online 8 July 2013

33/$ - see front matter & 2013 Elsevier Ltd. Ax.doi.org/10.1016/j.pss.2013.06.026

d on observations taken with the following tem and WHT 4.2 m in La Palma, OHP 1.2 m anurg 2 m and Bonn AIfA 0.5 m in Germany, Gania.esponding author at: Isaac Newton Group of T321, E-38700 Santa Cruz de la Palma, Canary4 92 2425461.ail addresses: ovidiu.vaduvescu@gmail.com,@ing.iac.es (O. Vaduvescu).

a b s t r a c t

We report follow-up observations of 477 program Near-Earth Asteroids (NEAs) using nine telescopes of theEURONEAR network having apertures between 0.3 and 4.2 m. Adding these NEAs to our previous results wenow count 739 program NEAs followed-up by the EURONEAR network since 2006. The targets were selectedusing EURONEAR planning tools focusing on high priority objects. Analyzing the resulting orbital improve-ments suggests astrometric follow-up is most important days to weeks after discovery, with recovery at a new

ll rights reserved.

lescopes: Blanco 4 m in Chile,d Pic du Midi 1 m in France,lati 0.4 m and Urseanu 0.3 m

elescopes (ING), Apartado deIslands, Spain.

O. Vaduvescu et al. / Planetary and Space Science 85 (2013) 299–311300

Keywords:Minor planetsNear Earth AsteroidsMain belt asteroidsAstrometry and orbitsFollow-up and discoverySurvey statistics

opposition also valuable. Additionally we observed 40 survey fields spanning three nights covering 11 squaredegrees near opposition, using the Wide Field Camera on the 2.5 m Isaac Newton Telescope (INT), resulting in104 discovered main belt asteroids (MBAs) and another 626 unknown one-night objects. These fields, plusprogram NEA fields from the INT and from the wide field MOSAIC II camera on the Blanco 4m telescope,generated around 12000 observations of 2000 minor planets (mostly MBAs) observed in 34 square degrees.We identify Near Earth Object (NEO) candidates among the unknown (single night) objects using threeselection criteria. Testing these criteria on the (known) program NEAs shows that the best selection method isour ϵ�μ model which checks solar elongation and sky motion and the MPC's NEO rating tool. Our new datashow that on average 0.5 NEO candidates per square degree should be observable in a 2 m-class survey (inagreement with past results), while an average of 2.7 NEO candidates per square degree should be observablein a 4 m-class survey (although our Blanco statistics were affected by clouds). At opposition just over 100 MBAs(1.6 unknown to every 1 known) per square degree are detectable to R¼22 in a 2 m survey based on the INTdata (in accordance with other results), while our two best ecliptic Blanco fields away from opposition lead to135 MBAs (2 unknown to every 1 known) to R¼23.

& 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Since Ceres was found by Piazzi in 1801, the discovery rate ofMain Belt Asteroids (MBAs) has increased, to the point where theIAU Minor Planet Center (MPC) has now cataloged over 600,000minor planets (Minor Planet Center, 2012a). Near Earth Asteroids(NEAs), defined as minor planets with perihelion distanceq≤1:3 AU and aphelion distance Q≥0:983 AU (Morbidelli andet al., 2002), represent an important Solar System population:their main formation mechanisms include migration of MBAs dueto resonances, especially 3:1 mean-motion with Jupiter and ν6with Saturn (Farinella et al., 1993), possibly combined with theYarkovsky/YORP effects (Bottke et al., 2006). There are more than9200 known NEAs today (Minor Planet Center, 2012a).

Potentially Hazardous Asteroids (PHAs) are defined as asub-class of NEAs having minimum orbital intersection dis-tance MOID≤0:05 AU and absolute magnitude H≤22 (Bowelland Muinonen, 1994). We know today more than 1300 PHAs(Minor Planet Center, 2012a). Virtual Impactors (VIs) are NEAswhose future Earth impact probability is non-zero according tothe actual orbital uncertainty (Milani and Gronchi, 2009). Thereare about 420 and 350 VIs listed in the NASA JPL Sentry RiskTable (NASA JPL, 2012) and NEODyS Risk List (NEODyS, 2012),respectively.

There is a continual need to observe asteroids in order to studytheir orbits and, for NEAs, to catalog future approaches to Earth.New astrometry improves the orbital quality which would other-wise be degraded by effects including close planetary approachesand non-gravitational effects (Yarkovsky, YORP).

The European Near Earth Asteroid Research (EURONEAR) wasinitiated in 2005 in Paris in order to bring some Europeancontribution to the NEA research using existing telescopes andhopefully some automated dedicated facilities (EURONEAR,2012a). Lacking dedicated funding, during recent years mostlyvolunteering students and amateur astronomers directed by a fewresearchers reduced the data in near real time (few hours or daysafter acquisition), some of them participating in observing runs.Vaduvescu et al. (2008) introduced EURONEAR and first observa-tions obtained at Pic du Midi in 2006, followed by Birlan et al.(2010a) who described the first 160 observed NEAs using ninetelescopes available to the EURONEAR network during the firstfour years. Vaduvescu et al. (2011) presented some MBA and NEAstatistics based on three runs using large field 1–2 m facilities(Swope 1 m, ESO/MPG 2.2 m and INT 2.5 m).

In this paper we present new NEA recovery and follow-upobservations using six professional and three educational-amateurtelescopes available to the EURONEAR network during the last twoyears. Adding these observations, the total number of observed

NEAs within the EURONEAR network reaches 739 objects (October2012). Thanks to two observing runs and a few discretionary timehours awarded to use the large field imaging cameras of the 2.5 mIsaac Newton Telescope (INT) and Blanco 4 m telescope, in a teamcomprised mostly of students and amateur astronomers weobtained and carefully analyzed 572 CCD fields covering 34 squaredegrees, visually scanning, measuring and reporting around 12thousand positions of more than two thousand asteroids. We havethen used the data obtained with 2–4 m telescopes to derive MBAand NEA statistics going beyond our previous results based on dataobtained with 1–2 m facilities (Vaduvescu et al., 2011).

Section 2 describes our observations. In Section 3 wecompile the results, including the classification into observedNEAs, known MBAs and unknown objects. In Section 4 wediscuss these results, focusing on the distribution of the knownand unknown MBAs and NEAs; we compare the INT and Blancofacilities and present statistics for the use of 2 m and 4 m classsurveys. Conclusions and two future projects are presented inSection 5.

2. Observations and data reduction

To distinguish between targeted and new candidate NEAs, wedefine as program NEA any Near Earth Asteroid (NEA, PHA or VI)programmed for follow-up within the EURONEAR network. Toplan our runs and select program NEAs given an observing placeand date, we used two planning tools (EURONEAR, 2012b) whichtarget the newly discovered objects from the Spaceguard prioritylist (Spaceguard System, 2012) and the NEA bright and faintrecovery opportunity lists maintained by the Minor Planet Center(Minor Planet Center, 2012a). More information about theseplanning tools can be found in Vaduvescu et al. (2011).

Ten cameras mounted on nine telescopes were used betweenJanuary 2009 and June 2012. In Table 1 we list specifications ofeach facility. We present each observing node and results next. Forall observing sites and runs except TLS we used Astrometrica(Raab, 2012) to identify program NEAs and other moving sources,using its field recognition, image registering and blinking capabil-ities, by visually scanning, measuring and reporting the observedfields in near real time (few hours or days after the runs). Forrelatively sparse observations and small sky area covered, Astro-metrica has been proven an excellent team capability for use bystudents and amateurs. Moreover, for limited amounts of data, thehuman eye and brain have been better for moving object detec-tion, reaching lower S/N detection versus automated pipelines (seefor example Vaduvescu et al., 2011 for some references andcomparison).

Table 1Telescopes and detectors used for follow-up, recovery and securing orbits of NEAs. The columns give the observatory, telescope aperture (m), CCD camera, number of CCDsand individual size in pixels, field of view and pixel scale.

Telescope Aperture Camera CCDs and Pixels FOV (′) Scale (″=pix)

ORM INT 2.5 WFC 4� (2048�4096) 34�34 L 0.333ORM WHT 4.2 ACAM 2048�4096 8 diam 0.254CTIO Blanco 4.0 MOSAIC II 8� (2048�4096) 36�36 0.27Pic du Midi (b) 1.05 iKon-L Andor 2048�2048 7.5�7.5 0.22Pic du Midi (c) 1.05 Atik 383L+ 3326�2504 7.8�5.8 0.14Haute Provence 1.20 TK1024 1024�1024 11.7�11.7 0.685TLR Tautenburg 2/1.34 SITe 2048�2048 42�42 1.23Bonn/AIfA 0.50 SBIG-STL 6303E 3072�2048 23.0�15.4 0.45Galati 0.40 SBIG STL-6303E 3072�2048 29.8�19.9 0.58Urseanu 0.30 TC273-home made 640�500 21.6�16.9 2.30

O. Vaduvescu et al. / Planetary and Space Science 85 (2013) 299–311 301

2.1. INT observing runs

We used the 2.5 m Isaac Newton Telescope (INT) owned bythe Isaac Newton Group (ING) at Roque de Los MuchachosObservatory (ORM) in La Palma. Four nights were awarded bythe Spanish Time Allocation Committee (CAT, 25–28 February2012, proposal number C6, PI: O. Vaduvescu) and approximately10 h were used in total during four service/discretionary INGnights(S/D) in 2011.

At the prime focus of the INT we used the Wide Field Camera(WFC) which consists of four CCDs 2K � 4K pixels each coveringan L-shaped 34′� 34′ with a pixel scale of 0:33″=pix. All frameswere observed with 2�2 binning (0:66″=pix) to decrease thereadout time, except for the December 2011 run which wasoperated without binning. During three of the S/D nights (exceptfor December 2011 run) we tracked the observed fields at half theNEA proper motion, while for the remaining runs we tracked fieldsat normal sidereal rate. The filling factor of the WFC (defined asthe ratio of the active area over total sky subtended camera area) is0.86. We used an R filter for all runs.1

The CAT time was dark, and the weather was good during threenights, except for the first 3 h during the second night and thethird night when we closed due to humidity. The typical seeing atthe INT was ∼1:5″. For data reduction we used THELI (Erben et al.,2005) to correct the field distortion in the prime focus of the INT,then using Astrometrica with a fit order 1–2 and USNO-B1 orNOMAD (thus UCAC2) catalogs.

Besides the program NEAs (listed in Table A1 in the Appendix),during three clear nights at the INT in February 2012 (the first,second and fourth of our run) we undertook an MBA mini-surveyaround opposition, covering in total 40 WFC fields (five successiveimages of the same field) during 12 h total time. The pointing gridof the MBA mini-survey was optimized in order to recover mostnew MBAs observed on previous night(s), assuming a dailyapparent motion of 17′ westward along the ecliptic (based on0:7″=min average speed). During the first night we observed 10arbitrarily chosen WFC fields close to opposition in the ecliptic.During the second night we observed the same 10 WFC fieldsshifted by �1 min in α and þ8′ in δ. During the fourth night weobserved a pair of fields shifted by �2 min in α and þ16′ and �1′in δ with respect to the second night. During the short timeavailable for this mini-survey, this pointing grid allowed us torecover many objects appearing in previous nights in the WFCfields and maximize the number of MBA discoveries.

Due to the large WFC field and INT aperture, we measured andidentified all moving sources detected visually via blinking in all

1 Historically we used R filter for most of our EURONEAR runs in order tominimize moonlight, artificial light pollution and also for photometric consistency.

observed fields in all INT runs, reporting all known and newobjects visible up to apparent magnitude R∼22. In total, at the INTwe observed 87 WFC fields covering ∼24 square degrees (includ-ing 40 WFC fields covering 11 square degrees in our opposi-tion mini-survey). We followed or recovered 33 program NEAs(including 4 PHAs and 1 VI), making the INT-WFC one of the mostproductive EURONEAR facilities for recovery and survey work(NEAs and MBAs). In Section 4.3 we use the INT findings forMBA and NEA statistics.

2.2. Blanco observing run

The first EURONEAR run using a 4 m telescope was awarded bythe Chilean Time Allocation Committee for 3–4 June 2011 (propo-sal number 0646, PI: E. Unda-Sanzana) to use the Victor Blanco4 m telescope at Cerro Tololo Inter-American Observatory (CTIO) inChile. At the prime focus F=2:9 of the telescope we employed theMOSAIC II camera which consists of a 2�4 mosaic of CCDs 2K � 4Kpixels each covering a total field of view of 36′� 36′ with a pixelscale of 0:27″=pix. To minimize the readout time and FTP datatransfer, we used 2�2 binning (binned pixel size 0:54″). We usedan R filter for the entire Blanco run. The filling factor of theMOSAIC II camera is 0.98. We used normal sidereal tracking for allfields.

The time was dark, with only the first night clear and its first3 h affected by clouds, resulting in no NEA recovered in the firstnine program NEA fields. The second night had complete cloudcover, one day before a big snow storm which hit Cerro Tololo.During the first night we recovered 12 program NEAs (including1 VI and 9 PHAs) plus two other NEAs falling by chance in theobserved fields (1 PHA). The typical seeing was ∼1″.

For data reduction we used the USNO-B1 catalog and no fieldcorrection due to lack of time to fit THELI to work in binning modewith MOSAIC II. First, we reduced data allowing the high fielddistortion to be accommodated by a fit order 3 in Astrometrica.Then, the MPC advised us about some errors (Williams, privatecommunication) which we traced to the high field distortion of theprime focus camera, thus we restricted Astrometrica to a fit order2 which allowed better astrometry.

Besides NEAs (Table A1), we identified and measured allmoving sources in all observed fields (except for four fields fallingin the Milky Way), reporting all known and new objects visible upto R∼24 mag. Using Blanco during one night, we observed in total28 MOSAIC II fields covering in total 10 square degrees. In Section4.3 we use Blanco data to derive some MBA and NEA statistics.

2.3. WHT observing runs

A few observations to recover some NEAs were conductedduring twilight or combined with some technical tasks during fourservice or discretionary (S/D) nights in 2011, using the William

O. Vaduvescu et al. / Planetary and Space Science 85 (2013) 299–311302

Herschel 4.2 m Telescope (WHT) of the ING located at ORM.Additionally, during other two service nights in 2011 we observedfor photometry the OSIRIX-Rex target (the VI (101 955) 1999RQ36) under the program SW2011a31 (PI: J. Licandro), for about8 h in total. In total using the WHT, we observed 11 program NEAs(including 4 PHAs and 1 VI). We used guiding only for thephotometry runs of (101 955), tracking at normal sidereal ratefor the other fields.

At the Cassegrain F=11 focus we used the ACAM camera whichconsists of one CCD with 2K � 4K pixels covering a field of view of8′ diameter with a pixel scale of 0:25″=pix. We used R filters for allWHT runs. The weather was mostly good, with WHT ACAM typicalseeing around 1″. In Table A1 we list the program NEAs observedwith WHT.

For data reduction we used IRAF for typical flat and biascorrections and Astrometrica with a fit order 1–2 (sufficient forthe Cassegrain instrument), using mostly USNO-B1, UCAC3 orUCAC4 depending on the star density in the observed smallACAM field.

2.4. TLS observing runs

Türinger Landessternwarte Observatory (TLS) located inTautenburg, Germany, joined the EURONEAR network in 2009,starting regular NEA observations in 2011 using the Alfred Jensch2 m telescope (1.34 m Schmidt corrector plate). At the Schmidt F=4focus of the TLS telescope is a SITe CCD 2k� 2k camera, covering alarge field of view 42′� 42′ with a pixel scale of 1:23″=pix.

During 39 nights in September 2011, March, and May 2012, atotal of 38 NEAs, mostly very recently discovered objects, wereobserved (Table A1). The median seeing in Tautenburg is ∼2″ andthe R filter was used for all runs. The TLS pipeline which includesrun planning and data reduction is complete, and few runs areplanned in the near future.

Astrometry and field correction are resolved using Astrometry.net (Lang et al., 2010) with the GSC 2.3 catalog. The exposure timeused at TLS was a standard 180 s for all NEAs. The observingtechnique and data reduction method are a novelty in EURONEAR,so we briefly present this here. Depending on the magnitude andproper motion of the object, a few frames are taken with sameexposure time in normal sidereal auto-guiding, thus the asteroidbecomes visible as a trail. After basic correction of bias and flatfield, field distortion and astrometry are resolved with astrometry.net, using second order SIP polynomials to account for fielddistortion. The median co-added image is created by registeringall images on the center of the NEA trail given the shifts calculatedfrom the ephemeris. Finally, the NEA trail will be deconvolved

Fig. 1. The very fast NEA 2012 FP35 observed at TLS using the trail deconvolution reductistacking the two individual frames. The right side presents the result after deconvolutioreferences to color in this figure caption, the reader is referred to the web version of th

using a PSF artificial trail (length and direction from motion, widthfrom stellar PSF) to model the centrally peaked elongated bright-ness distribution of the object. Finally, the co-added trail is fittedby an elliptical Gaussian to come up with object coordinates anderrors.

Fig. 1 shows the case of the extremely newly discovered (oneday) fast NEA 2012 FP35 moving with μ¼ 100:5″=min. The leftimage is the result of adding two registered frames, resulting intwo observed trails which do not exactly superimpose because ofthe large uncertainty in the orbit (e.g. position and proper motion).The predicted position is indicated by the green circle while theobserved one (the result of the fit) is marked by the red circle. Theright image represents the recovered deconvolved image wherethe stars become elongated.

2.5. OHP observing runs

Two observing runs (10 nights in total) were awarded in 20–24April 2010 and 15–19 November 2010 by the Local Time AllocationCommittee using the 1.2 m telescope (T120) at Haute de ProvenceObservatory (OHP) in France.

The Netwonian F=6 T120 telescope hosted the TK 1024 ABcamera equipped with a 1k� 1k CCD covering a 11:7′� 11:7′ fieldof view with a pixel scale of 0:68″=pix. The median seeing at OHPwas about 2:5″. The R filter was used for both runs and normalsidereal tracking. For data reduction we used Astrometrica with fitorder 1 and USNO-B1 catalog to provide enough stars for therelatively small and shallow observed fields.

Past (pre-2010) OHP observations were reported by Birlan et al.(2010a). In total during the 2010 runs we observed 36 programNEAs (including 8 PHAs; Table A1).

2.6. Pic du Midi observing runs

Two observing runs (10 nights in total) were awarded in 1–4March 2011 and 17–24 November 2011 by the Station de Planeto-logie des Pyrenees managed by Observatoire de Paris and Observa-toire Midi Pyrenees at Pic du Midi Observatory, France. For bothruns the T1m 1.05 m telescope was used.

For the first run, at the Cassegrain F=12:1 reduced focus weemployed the iKon-L Andor camera with a 2k� 2k E2V chip withpixel scale 0:22″=pix. To reduce the noise we used 2�2 binning, soa pixel size of 0:44″ with a field of view of 7:5′� 7:5′. For thesecond run at the Cassegrain F=7:6 reduced focus we employedthe Atik 383L+ camera with an 3326�2504 CCD Kodak KAF-8300chip with pixel scale 0:14″=pix. We used 3�3 binning, so a pixelsize of 0:42″ with a field of view of 7:8′� 5:8′. For both runs we

on method. On the left side we present the observed image obtained by shifting andn showing the data (magenta) and the fit (dark yellow). (For interpretation of theis article.)

O. Vaduvescu et al. / Planetary and Space Science 85 (2013) 299–311 303

used a clear filter from Astronomik equivalent to Bþ V þ R withbandpass from 390 nm to 680 nm, tracking at half the propermotion of observed NEAs.

Typical seeing at Pic du Midi was about 1:5″. The weather wasgood and the Moon was mostly dark. For the data reduction weused Astrometrica with fit order 1, enough for the Cassegrain field,and USNO-B1 and NOMAD catalogs, depending on available stardensity in the fields.

Past EURONEAR Pic du Midi observations were reported inVaduvescu et al. (2008) and Birlan et al. (2010a). During the twoPic du Midi 2011 runs, 34 NEAs were observed (including 5 PHAsand 1 VI; Table A1).

2.7. Bonn/AIfA observing runs

The Argelander Institute for Astronomy (AIfA) owns a 0.5 mtelescope hosting at its Cassegrain F=9 focus an SBIG-STL 6303ECCD camera (3072�2048 pixels, scale 0:45″=pixel). The typicalseeing there is about 3″ and the light pollution is quite high(naked-eye mag 4 at zenith) allowing mag 19 to be reached onlyafter stacking a few short exposed images (using Astrometrica's“track and stack” capability). We observed with 2�2 binningusing V and R filters, tracking all fields at normal sidereal rate. Weused Astrometrica with UCAC3 and USNO-B2 catalogs and no fieldcorrection in the Cassegrain field.

The MPC code of the Bonn/AIfA observatory C60 was obtainedby A. Tudorica (MSc student there) who joined the Bonn/AIfA nodeto the EURONEAR network in 2011. During 25 nights betweenSeptember 2011 and May 2012, we observed 60 NEAs (of which 15PHAs) and reported 830 positions.

2.8. Galati observing runs

The public Astronomical Observatory of the Natural SciencesMuseum Complex in the city of Galati, Romania, was found thanksto the efforts of O. Tercu, the actual observatory coordinator, basedon funding from the county of Galati (Consiliul Judetean Galati),plus other funding from the E.U. and some local sponsorship. Theobservatory joined the EURONEAR network in 2010 and obtainedthe MPC code C73 in 2011. Supported by a small but enthusiasticteam of amateurs in the local astronomy club “Calin Popovici”,Galati observatory includes NEAs as one of their main priorities forpublic outreach and education.

Among other smaller telescopes and instruments, the maintelescope of Galati observatory is the ASA 0.4 m Ritchey–ChretienF/8 telescope endowed with an SBIG STL-6303E CCD with 3072�2048 pixels covering a relatively large field of view 29:8′� 19:9′with pixel scale 0:58″=pix.

All NEA observations of Galati Observatory were performedusing 2�2 binning (binned pixel size 1:16″=pix), due to the typicalseeing of about 3″. For all runs we used normal tracking at siderealrate. For data reduction we used Astrometrica with UCAC3 catalogand a fit order 4 to correct the relatively distorted large field.

During 55 nights between September 2011 and May 2012 threepeople observed 223 NEAs (of which 40 PHAs and 4 VIs) andreported 2367 positions, becoming the most prolific EURONEARsite in terms of number of observed objects and NEA publicoutreach. Despite some city light pollution, the 40-cm telescopemanages to reach limiting magnitude V∼19.

2.9. Urseanu observing runs

The public “Admiral Vasile Urseanu Observatory” in Buchar-est, Romania, was found in 1910 by the navy admiral VasileUrseanu who was also an amateur astronomer. The actualpublic meeting place of many visitors and host of the Bucharest

Astroclub, the observatory is owned by the Bucharest Museumof History. Thanks to the efforts of A. Sonka, in 2006 theobservatory obtained its MPC code A92, becoming one of thefirst EURONEAR nodes.

The very small 0.3 m telescope (Meade LX200R) of the UrseanuObservatory is actually the smallest used by our network, beingextensively used for public outreach there. At the Cassegrain F/6.3field resides an inexpensive QHY6 CCD camera. The pixel scale is2:30″=pix and the field is 21:6′� 16:9′. Typical seeing in the verylight polluted capital Bucharest using this equipment is about 3″and the limiting magnitude is V∼16 (V∼15 for fast movingobjects). For all runs we used sidereal tracking.

Past NEA observations from Urseanu Observatory werereported in Vaduvescu et al. (2008) and Birlan et al. (2010a). FromNovember 2008 until May 2012 a total of 28 NEAs (including9 PHAs and 1 VI) were observed and reported from thisobservatory.

3. Results

3.1. Program NEAs

A total of 477 program NEAs observed within the EURONEARnetwork between 2009 and May 2012 are reported in this paper.Of these, 166 were observed with six professional 1–4 m tele-scopes and 311 with three smaller (0.3–0.5 m) educational andpublic outreach telescopes. In the Appendix Table A1 we give theobserving logs containing these objects observed with the abovenine telescopes. In Fig. 2 we plot the O�C (observed minuscalculated) residuals in α and δ for the program NEAs observedwith each telescope.

The root mean square of the O�C residuals for our MPCpublished NEA datasets for each telescope are in the order: WHT0:44″, INT 0:46″, Blanco 0:46″, Bonn 0:56″, Galati 0:57″, Pic 0:81″,OHP 0:85″, Urseanu 1:05″ and TLS 1:14″.

Overall, small amateur-educational telescopes Bonn and Galatiare performing very good astrometry, only slightly inferior by 0:1″than their much larger 2–4 m “colleagues” INT, Blanco and WHTwhich also benefit from much better weather conditions. Owing toless accurate catalogs used (USNO-B1 versus UCAC3) and possiblydue to target selection (newly discovered versus new oppositionobjects) we observe that O�Cs of our two 1 m facilities OHP andPic trail behind Bonn and Galati by about 0:3″. Thus, the loss inastrometric precision is absolutely non-linear with the shorteningof aperture, the most important factors favoring the contributionfrom small telescopes being better astrometric catalogs used(especially in the bright regime more accessible to small tele-scopes), a good SNR for both target and reference stars and theabsence of complicated field distortion, all in line with recom-mendations of the IAU Working Group “Astrometry by SmallGround Based Telescopes”.

At least two inhomogeneities can be spotted in the O�Cplots in Fig. 2. First, the small vertical branch in the OHP plot isdue to few OHP objects moving much faster in δ than in α,resulting in larger measurement errors and O�Cs in δ. Second,the WHT O�C centroid seems slightly displaced (by about 0:2″)to the upper-right from the origin, which is explained by theknown systematics of the mostly used USNO-B1 catalog due tothe small ACAM field (a fact observed also by us previously,Birlan et al., 2010a).

3.1.1. MPC/MPEC publicationsAdding the 477 observed NEAs to our previous work presented

in Birlan et al. (2010a) and Vaduvescu et al. (2011), we report atotal of 739 NEAs observed within the EURONEAR network during

Fig. 2. O�C (observed minus calculated) residuals for program NEAs observedusing the telescopes presented in this paper.

O. Vaduvescu et al. / Planetary and Space Science 85 (2013) 299–311304

the last six years. The reduced data presented in this papergenerated 98 MPC publications, comprising 69 Minor PlanetCirculars (MPCs) and 29 Minor Planet Electronic Circulars (MPECs).

The Blanco run resulted in published data in the following eightMPC/MPEC publications: 74 496, 75 198, 75 623, 75 939, 76 865,79 528 (Buie et al., 2011a,b,c,d,e, 2012), 76 441 and 77 265 (Elstet al., 2011a,b).

The INT runs resulted in published data in the following 11MPC/MPEC publications: 76 443, 78 047, 78 437, 78 894, 79 221,79 530 (Fitzsimmons et al., 2011a, 2012a,b,c,d,e), 75 625 (Holmanet al., 2012), 79 787 and 2012-D82 (Vaduvescu et al., 2012a,b),2012-E19 (Gajdos et al., 2012) and 2012-D102 (Bressi et al., 2012).

The WHT observations resulted in published data in thefollowing eight MPC/MPEC publications: 74 148, 74 893, 75 207,76 443, 77 266, 78 047 (Fitzsimmons et al., 2011a,b,c,d,e, 2012a),75 625 (Holman et al., 2012) and 75 940 (Balam et al., 2011).

The OHP runs resulted in published data in the following six MPC/MPEC publications: 69 732, 72 456 (Arlot et al., 2010a,b), 70 198,2010-W11, 2010-W12 and 2010-W13 (Birlan et al., 2010b,c,d,e).

The Pic du Midi runs resulted in published data in the following17 MPC publications: 74 036, 77 173 (Cavadore et al., 2011a,b),2011-E11, 2011-E12, 2011-E13, 2011-W29, 2011-W33 (Birlan et al.,2011a,b,c,d,e), 2011-E14 (Lehmann et al., 2011), 2011-E19 (Apitzschet al., 2011), 2011-W25 (McMilan and et al., 2011), 2011-W12,2011-W22, 2011-W27, 2011-W28, 2011-W44, 2011-W45 and 2011-W52 (Buzzi et al., 2011a,b,c,d,e,f,g).

Tautenburg runs resulted in published data in the following sixMPC/MPEC publications: MPC 79 140, 79 473 (Borngen andStecklum, 2012a,b), 79 746, 79 971 (Stecklum, 2012a,b), 2012-G45and 2012-K07 (Stecklum et al., 2012a,b).

Bonn AIfA runs resulted in published data in the following nineMPC/MPEC publications: MPC 74 509, 79 224, 79 789, 80 000(Tudorica, 2011, 2012a,b,c), 76 451 (Tudorica and Toma, 2011),76 874, 78 902 (Tudorica and Badescu, 2011, 2012), 77 273 and79 533 (Tudorica et al., 2012d,e).

The Galati runs resulted in published data in the following 17MPC/MPEC publications: MPC 75 947, 76 451, 76 874, 77 273, 77 705,78 053, 78 445, 78 902, 79 225, 79 533, 80 000 (Tercu and Dumitriu,2011a,b,c,d, 2012a,b,c,d,e,f,g), 2012-B17 (Bacci et al., 2012), 2012-G45(Stecklum et al., 2012a), 2012-H42 (Eglitis et al., 2012), 2012-H90(Jaeger et al., 2012), 2012-J01 (Nishiyama and et al., 2011) and 2012-J34 (Christensen et al., 2012).

Finally, Urseanu runs resulted in published data in the follow-ing 16 MPC/MPEC publications: MPC 64 104, 65 048, 65 638,65 928, 67 139, 67 404, 67 677, 698 17, 71 094, 72 053, 75 208,75 447, 77 270, 77 702, 78 050 and 79 998 (Sonka, 2008, 2009a,b,c,d,e,f, 2010a,b,c, 2011a,b,c, 2012a,b,c).

3.1.2. Orbital improvementWe focus next on the NEAs recovered at a new opposition (29

objects marked by ⋆ in Appendix Tables A1 and A2) and also onthe NEAs followed-up soon after discovery—new objects whosearcs were either observed within the first day or prolonged by atleast one day during the first month (82 objects marked by � inTables A1 and A2), studying the orbits of these 111 objects.

We use the software FIND_ORB (Gray, 2012) to assess theorbital improvement, by comparing the orbital elements fittedwith observational data available before our runs (first line inTable A2) with those obtained adding our observations (secondline in Table A2). We list in the table the semimajor axis a,eccentricity e, inclination i, longitude of ascending node Ω, argu-ment of pericenter ω, mean anomaly M and minimum orbitalintersection distance MOID.

Despite the importance of the recovery of NEAs at a newopposition which lengthens the observed arc by a few years,orbital elements of the NEAs recovered at a new opposition appearto be improved only slightly in absolute terms: standard devia-tions of the changes to elements (based on 25 pairs of orbits for

O. Vaduvescu et al. / Planetary and Space Science 85 (2013) 299–311 305

objects marked ⋆ in Table A2) are a∼10�4 AU, e∼5� 10�5,i∼10�3 deg, Ω∼10�3 deg, ω∼3� 10�3 deg, M∼5� 10�2 deg, MOID∼3� 10�5 AU and s∼0:02″ (RMS of the fit).

Orbits of NEAs followed soon after discovery, as they are lesswell constrained a priori, naturally undergo larger changes to theelements following our observations, with standard deviations ofa∼10�1 AU, e∼10�2, i∼1 deg, Ω∼1 deg, ω∼1 deg (all about 103

times more), M∼7 deg, MOID¼0.0151 AU and s∼0:06″, based on73 pairs of orbits representing objects observed with all telescopesmarked � in Table A2.

FIND_ORB additionally provides the option to calculate uncer-tainties in the orbits it determines, enabling us to assess the orbitalquality both before and after our observations (as opposed tosimply the change in the best estimate orbit due to our observa-tions). We therefore calculated 1-sigma a, e and i uncertainties sa,se and si for Table A2 objects and can define orbital improvementas the ratio in the sa, se or si values before and after ourEURONEAR observations. Fig. 3, and similar plots for e and i notshown, suggests that the orbital improvement is closely related tothe increase in time interval spanned by the observations. This canbe quantified by fitting log of orbital improvement to log ofproportional increase in timespan. The formal linear regressionresults were

Fig.aboof eincrclar

log ðsa ratioÞ ¼ ð1:3570:22Þlog T þ ð0:2070:15Þ ,log ðse ratioÞ ¼ ð0:8170:19Þlog T þ ð0:2670:13Þ,log ðsi ratioÞ ¼ ð0:5770:15Þlog T þ ð0:2170:10Þ,

where T is the ratio of observed timespan after to before, and 7refers to 95% confidence intervals. Considering separately theobjects observed in the days to weeks after discovery, the sa, seand si regression slopes were all close to 1.0. For the objectsrecovered at a new opposition, the slope was 1.0 for sa, and 0.6 forse and si (but 1.0 marginally permitted at 95% confidence forboth). A value of 1.0 would imply that the proportional orbitalimprovement equals the proportional increase in observed time-span. In practice, the number and distribution of observations

3. Orbital improvement as a function of observational timespan. Distanceve diagonal line corresponds to factor by which observed arc has increased. Sizeach circle shows factor by which uncertainty in a has improved as a result ofease in observed arc. Points near (1 day, 1 day) have been slightly separated fority.

within the timespan, and the accuracy of the observations, alsoaffect the orbital quality (cf. Muinonen et al., 1994).

Based on the analyzed datasets, we conclude that the mostimportant orbital contribution is the rapid follow-up of newlydiscovered NEAs (few days to one month after discovery).Another contribution is the recovery of NEAs at a new opposi-tion which enlarges orbital arcs by a few months or years.Obviously, VIs and PHAs have priority over regular NEAs. Mostimportant are fainter objects (V421) which could remaininvisible to the available 1–2 m surveys and could result ininsecure orbits or lost objects not being observed for the nextfew years.

3.2. Known MBAs

Many known MBAs were serendipitously encountered, identi-fied, measured and reported in the large field images taken byBlanco MOSAIC II and INT WFC cameras. In total, 1699 positionscorresponding to 288 known MBAs were reported during ourBlanco Mosaic II run, while 3465 positions of 580 known MBAswere reported from the observed INT-WFC fields.

In Fig. 4 we plot the O�C astrometric residuals correspondingto our reported positions of the known MBAs encountered in allBlanco and INT fields. The first panel plots the residuals for theBlanco run clearly affected by the field distortion of the MOSAIC IIcamera, with a root mean square of the O�C residuals 0:90″ andsome values larger than 2″ for the objects measured far from thecenter of the camera. In the second panel we plot the residuals forthe INT-WFC fields observed during December 2011 and February2012 runs (3465 positions). All the INT-WFC images were reducedwith THELI software (Erben et al., 2005) which corrected the fielddistortion across the whole WFC field. The INT (right) O�C plot inFig. 4 shows the accuracy of the WFC astrometry improved afterfield correction (RMS of O�C residuals 0:41″). By comparison, asimilar INT-WFC plot in our previous paper (Fig. 4b fromVaduvescu et al., 2011) includes WFC data processed withoutTHELI and shows an RMS more than double around the origin(0:97″).

Some inhomogeneity could be observed comparing theINT NEA O�C plot (Fig. 2) with the INT MBA plot (Fig. 4), namelythe NEA plot seems more elongated towards the α direction whilethe MBA plot is more symmetric in both α and δ. This is explainedby the fact that residuals for NEAs are larger than for MBAsdue to the measurement errors caused by NEAs moving faster in αthan in δ.

3.3. Unknown objects

Working in a team, we could detect visually and measure allmoving objects appearing in all fields observed with the INT andBlanco telescopes. All CCD images were bias corrected and flatfielded using IRAF or other image processing software, and onlythe WFC images were corrected for field distortion with THELIsoftware (Erben et al., 2005). We used Astrometrica (Raab, 2012)to identify all moving sources in the WFC and MOSAIC II pairedCCDs based on the MPCORB asteroid database retrieved soon afterthe observing date. We measured and reported all the unknownobjects which were later characterized by determining prelimin-ary orbits using the FIND_ORB software (Gray, 2012) in batchmode. Their main orbital parameters (a, e, i and MOID) are given inthe Appendix Tables A3–A7. Based on the observed fields and runs,we distinguish the following lists which total 1090 previouslyunknown asteroids:

�

104 unknown objects discovered mostly during multiple nightsin the February 2012 WFC opposition fields (Table A3);

Fig. 4. O�C (observed minus calculated) residuals for known MBAs observed with Blanco and INT. INT-WFC images were corrected with THELI for field distortion,decreasing residuals across the whole field to RMS standard deviation 0:41″.

O. Vaduvescu et al. / Planetary and Space Science 85 (2013) 299–311306

�

626 unknown objects observed with the INT during one nightin February 2012 in the opposition mini-survey fields(Table A4);

�

89 unknown objects observed with the INT in February 2012 inprogram NEA fields (Table A5);

�

75 unknown objects observed with the INT during 2011 runs inprogram NEA fields (Table A6);

�

1.3

2.0

3.5

1.3

2.0

3.5

Fig. 5. Using the ϵ�μ orbital model (Vaduvescu et al., 2011) to check NEOcandidates based on direct observational quantities solar elongation (ϵ) and propermotion (μ). We plot with solid symbols all unknown objects observed with Blanco(green) and the INT (red for objects observed in program NEA fields and blue forobjects observed in opposition fields). The three overlaid dotted magenta curvescorrespond to asteroids orbiting between a¼2.0 and a¼3.5 AU (main belt) anda¼1.3 AU (NEO limit). We mark with circles the 18 best NEO candidates labeled inbold in the Appendix Table A8. The fast NEO candidate ESU031 (identified with theknown NEA 2012 DC28) is located above the plot, matching the model. With blackcrosses we plot the 47 program NEAs (which include 11 fast objects located abovethe plot), most of them (38 objects or 81%) matching the ϵ�μ model. (Forinterpretation of the references to color in this figure caption, the reader is referredto the web version of this article.)

196 unknown objects observed with Blanco telescope duringone night in 2011 in program NEA fields (Table A7).

3.3.1. Discovered MBAs104 unknown objects could be recovered during multiple

nights in the February 2012 WFC opposition mini-survey, thusthey have been given provisional designations and should soonbecome credited MBA discoveries of our EURONEAR programusing the INT WFC. For these 104 discovered objects, AppendixTable A3 lists firstly their orbital elements a, e, i and MOID(calculated with FIND_ORB in batch mode) based on observationaldata from our run only (linked by the MPC), and secondly theirpublished MPC orbits based on all available observations in theMPC database (accessed 8 August 2012). The calculated FIND_ORBorbits are similar to those published by the MPC for the majority ofour discovered MBAs, with standard deviations 0.26 AU in a, 0.14in e, 3.78 deg in i and 0.30 AU in MOID. In fact the MPC orbits arecalculated mostly using our data alone, but in some cases even-tually adding other observations from elsewhere before or follow-ing our runs.

3.3.2. Unknown MBAsThe great majority of the 1090 unknown objects detected in the

INT and Blanco runs could be characterised as MBAs, their propermotion and preliminary orbits matching the known main beltpopulation well. We include the remaining unknown objects inAppendix Tables A4–A7, listing orbital elements a, e and i obtainedby FIND_ORB automated fitting of our observational data. Due tothe very short arcs derived from one night observations, theFIND_ORB preliminary orbits should be regarded with caution.Nevertheless, FIND_ORB was found to fit most orbits quite accu-rately, following direct comparison of one night fitted orbits withpublished orbits based on all available observations, and also aftercomparison with the whole known asteroid population via theclassic a�e and a�i plots (Figure 7 of Vaduvescu et al., 2011) whichcan be virtually reproduced using the 1090 preliminary orbitsfound here.

3.3.3. NEO candidatesFollowing our previous work which analyzed data taken with

1–2 m telescopes (Vaduvescu et al., 2011), we use three indepen-dent search methods to check the unknown objects for Near EarthObjects (NEOs) observed with 2–4 m facilities. The first methodemploys our model presented by Vaduvescu et al. (2011) whichplots the two directly observed quantities μ (apparent propermotion) and ϵ (solar elongation, ϵ above 180 deg corresponding tosky directions east of opposition). Using this model (Fig. 5), onecan distinguish NEO candidates (located above the a¼1.3 AU

Fig. 6. Number of unknown objects as a function of observed apparent R magnitude and calculated absolute magnitude H for the INT dataset (red solid line for NEA fields,blue dotted line for opposition fields) and Blanco (green dashed line for NEA fields). (For interpretation of the references to color in this figure caption, the reader is referredto the web version of this article.)

O. Vaduvescu et al. / Planetary and Space Science 85 (2013) 299–311 307

dotted magenta curve corresponding to the NEO limit) from MBAobjects (located below the a¼2.0 AU curve). The last column ofAppendix Table A8 lists the result of this fit as “Best”, “Good”,“Close” or “Bad” with respect to this model, considering onlyobjects flagged “Good” or “Best” as NEO candidates.

Our second NEO search method uses the “NEO Rating” serverdeveloped by the MPC (2012b) which calculates a score for NEOcandidates based on their expected proper motion compared tothe MBA distribution. In the second last column of Table A8 weinclude the MPC rating on a scale from 0% (worst) to 100% (bestfit), considering objects having scores higher than 10% as NEOcandidates.

Our third NEO search method uses the MOIDs of the prelimin-ary orbits calculated with FIND_ORB. It should be regarded withcaution because of the small observed arcs, but can indicatepossible NEOs. We take MOID o0:3 AU as the criterion for NEOcandidates (Table A8).

From 1090 unknown objects observed with INT and Blanco(Tables A3–A7), we find 60 NEO candidates defined as satisfying atleast one of the three search methods (Table A8). To allow a saferselection, we highlight 18 best NEO candidates defined as unknownobjects matching at least two of the three selection methods (boldin Table A8 and marked with circles in Fig. 5).

Five of these 18 best NEO candidates were observed with theINT in 2012 and 2011 in program NEA fields: EBA012, EPA015,VFLLP01, VKF008 and VKF034. The first three fulfilled all threeNEO search criteria. EBA012 was quite a fast moving object(μ¼ 2:13″=min) observed in the field of NEA 2012 BB14 (recoveredand published). EPA015 was a fairly bright object (R¼19.6)observed in the field of the NEA 2011 FR29 (recovered andpublished). It too was among the fastest of the best NEO candi-dates (μ¼ 3:33″=min), which classifies it as a NEA. VFLLP01 was arelatively faint object (R¼21.7), encountered at 12 September 2011as the only object in the field of PHA 2011 BN24 (not found atV¼21.6), although the EURONEAR O�C calculator does not showany correlation between the two objects. Its FIND_ORB fit is veryunstable based on the least squares, Herget or downhill simplexmethods, thus its calculated orbit should be regarded with caution.

Six best NEO candidates were observed with the INT during2012 in the opposition mini-survey: EBA023, EPA143, EPA190,EPO031, EPO065 and ESU031 (Table A8). ESU031 represents ourfastest NEO candidate at μ¼ 10:32″=min, producing long trails inthree 180 s exposures taken in the opposition OP5 mini-survey

field observed during the first night 25 February 2012. Wepromptly reduced and submitted three positions to MPC, but therobot matched it with 2012 DC28, an NEA discovered only twodays before by the Catalina survey. EPO031 fulfils all three NEOselection criteria and has relatively high brightness (R¼19.6): theMPC linked it with an object seen by Catalina, designating it 2012DL54, but its orbit is still indeterminate.

The remaining seven best NEO candidates were observed withthe Blanco telescope: PCTV024, PCTV026, PCTVb50, PCV023,PCVP005, PCVP007 and PCVS024 (Table A8). The first two wereencountered in the field of PHA 2008 YS27 (not found). PCTV024 isthe brightest (R¼18.3) of our 18 best NEO candidates, with MPCscore 100% and estimated MOID close to the NEA limit (0.28 AU).PCTV026 is a very similar object measured only in three images.However, both objects give a ϵ�μ “Bad” flag, while their measuredmagnitudes are highly uncertain (despite their brightness), beingaffected by the very high Milky Way star density seen with the 4 mtelescope. PCV023 is a relatively fast object μ¼ 1:78″=min and, asthe only one of the seven Blanco objects meeting all three NEOsearch criteria, is among the most promising of all our best NEOcandidates.

4. Discussion

4.1. Comparison between NEO search methods

At present (October 2012) 16 of the 18 best NEO candidates(Section 3.3.3) remain as one-night objects, with one (2012 DC28)confirmed as a certain NEA. We test the suitability of the threeNEO search criteria by applying them to objects known to be NEOs,namely the 47 program NEAs observed with INT and Blanco. Wetest the search methods using our observed small arcs only(knowing the objects to be NEOs from the MPC published orbitsbased on all available observations).

To test the ϵ�μ model, in Fig. 5 we plot with crosses all 47program NEAs. About 38 objects (81%) are confined above the NEOlimit, while nine outliers have very elliptical orbits, namely 2008QT3 (e¼0.52), 2011 AG5 (e¼0.39) and 2010 XC25 (e¼0.53) atbottom-left of the plot, 2011 XZ1 (e¼0.46) and 2009 OG (e¼0.86)close to opposition, and 2008 XB1 (e¼0.37), (175 706) (e¼0.35),(175 189) (e¼0.39) and 2007 JF22 (e¼0.59) at bottom-right ofthe plot.

Table 2Summary and statistics based on the Blanco and INT runs.

Observations Blanco INT Total

Recovered NEAs 14 33 47Nr. of positions 94 221 315

O�C standard deviation (″)… program NEAs 0.46 0.46 –

… known MBAs 0.97 0.41 –

Known MBAs 288 580 868… in NEA fields 288 132 420… in opp. fields – 448 448Nr. of positions 1699 3465 5164

Unknown objects 196 790 986… in NEA fields 196 164 360… in opp. fields – 626 626Nr. of positions 1301 4025 5326

Discovered objects – 104 104Nr. of positions – 1153 1153

Nr. of observing nights 1 4 5Nr. of program NEA fields 28 47 75Nr. of opp. fields – 40 40Total nr. of fields 28 87 115Total nr. of CCD images 224 348 572

Sky coverage (sq. deg) 10 24 34… NEA fields (sq. deg) 10 13 23… opp. fields (sq. deg) – 11 11

Limiting magnitude (R) 24.0 22.0 –

Apparent magnitude peak (R) 22.5 21.0 –

O. Vaduvescu et al. / Planetary and Space Science 85 (2013) 299–311308

Using the same 47 program NEA sets of observations, the NEOrating tool has 36 of them (77%) satisfying the recommended MPCInt450% limit (MPC, 2012b). Of the remaining 11 objects, 8 alsofail the ϵ�μ model, namely: 2008 QT3 (Int score 15%), 2010 XC25(25%), 2011 XZ1 (3%), 2009 OG (17%), 2008 XB1 (20%), (175 706)(score 23%), (175 189) (27%) and 2007 JF22 (score 4%). Three otheroutliers are 2009 EE81 (10%), 2006 CT10 (16%) and 2012 AC13 (rate47%). Only two objects have a NEO score below 10%, supportingthis limit as a safe threshold for our selection method of NEOcandidates.

Running FIND_ORB for the same 47 program NEAs, using onlyour short observed arcs, MOIDs of 11 objects result larger than theNEO limit of 0.3 AU, i.e. 36 (76%) would be NEOs according to thistest. Nevertheless, comparing these estimated MOIDs to the moreaccurately known MOIDs from MPC published orbits based on allavailable observations, the standard deviations are quite large,namely 0.90 AU in a, 0.34 in e, 12:9 deg in i and 0.28 AU in theMOID, thus the values of the FIND_ORB fitted orbits of NEAs basedon very small arcs should be regarded with great caution.

In comparison the same three search criteria applied to ourunknown objects resulted in 13 objects as NEO candidates accord-ing to the ϵ�μ model (flags “Best” or “Good”), 26 objects accordingto the MPC “NEO rating” selection (Int score 410%), and 41objects according to their FIND_ORB fitted orbits (MOID o0:3 AU),with 18 best NEO candidates satisfying at least two of thesecriteria (Section 3.3.3). Based on this data sample, the ϵ�μ modelclosely followed by the NEO Rating appears to be the safestmethod to flag one-night NEO candidates.

Total nr. of objects 498 1507 2005Total nr. of positions 3094 8864 11,958

Known MBA density (obj/sq. deg)… in NEA fields 29 10 –

… in opp. fields – 41 –

Unknown MBA density (obj/sq. deg)… in NEA fields 19 13 –

… in opp. fields – 66 –

Unknown NEO density (obj/sq. deg) 0.7 0.5 –

4.2. Comparison between 2 m and 4 m facilities

The Blanco 4 m run consisted of only one dark clear night,partially affected by thin clouds. The INT runs consisted of aboutfour mostly dark clear nights in all, three during the February 2012run, and time totalling about one night during four nights in 2011.Thus, the INT 2.5 m data could be more representative statisticallythan the Blanco 4 m data.

In Fig. 6 we plot histograms counting the unknown objects as afunction of the observed R magnitude (left) and calculatedabsolute magnitude H (right) based on the three available data-sets: the INT NEA fields (2011 and 2012), INT opposition fields(2012), and Blanco NEA fields. According to the left plot, the Blancotelescope sampled about 2 mag deeper than the INT, being mostefficient in detection around R∼22:5 and reaching a limit of R∼24.The INT was most efficient around R∼21 and reached a limit R∼22.Based on the left tail of the total distribution, apparently there areno more unknown objects brighter than R∼19 to discover nowa-days. The INT limit is higher by about 0.8 mag than our past results(R∼21:2 in Vaduvescu et al., 2011), probably due to better weatherconditions in 2011–2012. Also, the INT limit appears to surpass byabout 0.5 mag our past dark time ESO/WFI limit (R∼21:5). The Hhistogram suggests that the Blanco telescope is finding a higherproportion of larger (brighter H) objects than the INT, presumablybecause many of these unknown objects are at larger distances(thus fainter R), i.e. Blanco tends to see further out into themain belt.

Table 2 summarizes data derived from Blanco and INT runs, inobserved NEA fields and opposition (opp.) fields. Based on theO�C residuals, the astrometric quality for program NEAs observedmostly in the center of the fields is identical for the two cameras,with a root mean square of the O�C residuals of 0:46″. Never-theless, the astrometry of the known MBAs recovered across thewhole MOSAIC II field is more than two times worse (RMS 0:97″)due to the uncorrected field of Blanco at its prime focus, comparedwith the INT WFC (RMS 0:41″).

Counting the unknown and known asteroids (mostly MBAs)encountered in the Blanco and INT fields, we could assess theratio of the density of unknown to known MBAs in any 4 m and2 m class survey. The 87 observed WFC fields give this ratio forthe INT as 1.6 in opposition fields and 1.2 in program NEA fields,thus an average of 1.4. This fraction is very consistent with the1.3 result based on ESO/MPG data from our previous work and ishigher than our past 0.8 result based on previous INT dataaffected by bad weather (Vaduvescu et al., 2011). Counting theunknown and known objects observed with Blanco during onenight in 28 fields, the ratio is 0.7 for Blanco, unexpectedlysmaller than that from the INT. This inconsistency could beexplained due to poor statistics affected by weather and fieldselection. During the first part of the only Blanco night theclouds and cirrus strongly affected 9 fields from the total of 28observed. Four other fields were in the Milky Way and could notbe scanned for objects (other than the program NEAs) due to theextreme star density there. Also, seven other fields wereobserved at high ecliptic latitude and could not serve forstatistics. Eliminating all these fields, we selected five fieldswith limiting detection magnitude R∼23 (2008 XB1, 2008 DJ,2008 EP6, 2010 XC25 and 2009 CS), more suitable for statistics.Based on this selection, we derive the ratio 2.7 for Blanco inrelatively good weather conditions. In conclusion, Blanco coulddiscover about twice as many unknown MBAs with respect toknown MBAs (ratio 2.7) compared with the INT (ratio 1.4),

O. Vaduvescu et al. / Planetary and Space Science 85 (2013) 299–311 309

assuming similar good weather and dark sky conditions for bothfacilities.

4.3. Statistics of the unknown MBA and NEO population

Next we derive some statistics based on the data obtained from10 square degrees covered by Blanco and 24 square degreescovered by the INT (11 square degrees near opposition and 13square degrees in random NEA fields).

4.3.1. Unknown MBA densityAn average of 29 known MBAs and 20 unknown MBAs per

square degree could be observed with Blanco based on our smallanalyzed sample (mostly far from ecliptic) partially affected bypoor weather. The best two fields observed in good conditions inthe ecliptic (2010 XC25 and 2009 CS, both far from opposition)give similar densities to each other and an average of 45 knownMBAs and 90 unknown MBAs per square degree, thus a total of135 MBAs per square degree visible with the Blanco4 m telescope(two unknown to every one unknown asteroid) up to R∼23. Thisdensity based on small non-opposition statistics is comparablewith the work of Yoshida and et al. (2003) who counted 208 MBAsper square degree up to R∼23:0 based on the opposition SMBAS-Isurvey and 182 MBAs per square degree to the same limit from theSMBAS-II survey (Yoshida and Nakamura, 2007).

Based on our larger INT dataset, 41 known and 66 unknownMBAs per square degree could be detected in average in theobserved fields. Thus, a total of 107 MBAs per square degree (1.6unknown to every1 known objects) could be detected at oppositionby the INT up to R∼22:0. This density is comparable with ourprevious result of 63 MBAs per square degree (27 known, 36unknown) based on a similar survey using ESO/MPG (Vaduvescuet al., 2011) up to a limit 0.5 mag shallower. The SMBAS surveyswhen counted to the same R¼22.0 limit give 128 objects persquare degree (Yoshida and et al., 2003) and 103 objects persquare degree (Yoshida and Nakamura, 2007), very close to ourpresent INT result.

4.3.2. Unknown NEA densityConsidering the best NEO candidates marked in bold in the

Appendix Table A8, we estimate the unknown NEO densitydetectable in 2 m and 4 m surveys. Counting six NEO candidatesin the INT NEA fields covering 13 square degrees, we derive anaverage of 0.5 NEOs per square degree observable with the INT. Thisis in very good agreement with five best NEO candidates observedin the INT opposition fields covering 11 square degrees and alsowithin the range of our previous results (between 0.2 and0.6 objects per square degree based on the ESO/MPG and from0.1 to 0.8 for the INT, Vaduvescu et al., 2011). In the Blancoprogram NEA fields covering 10 square degrees, we count sevenbest NEO candidates, thus 0.7 NEOs per square degree observablewith the Blanco telescope; this number should be regarded as aminimum, based on poor weather during the only availableBlanco night.

The MPC lists 609,956 known asteroids and 9242 known NEAsin its database (checked 25 October 2012), a ratio of 1 in 66. OurNEO candidate list (Table A8) comprises 60 NEO candidates fromwhich only 18 are the best NEO candidates. Taking into accountour total of 1090 unknown objects observed with Blanco and INT,we derive a ratio of one NEO candidate for every 18 asteroids andone best NEO candidate for every60 unknown asteroids. This ratiocomes very close to the above known MPC ratio of 66, validatingour NEO candidate selection method.

4.3.3. Total number of NEAs observable by a 2 m surveyBased on the 11 best NEO candidates discovered in 24 square

degrees surveyed with the INT, we can evaluate the total numberof unknown NEOs detectable by a 2 m-class telescope. Checkingthe orbital distribution of our NEO candidates (11 INT objects inbold in the Appendix Table A8) we observe that all these objectshave inclination io251 (although the uncertainties are expectedto be large due to small observed arcs). Considering the wholeknown NEA population (NEODyS 2012), we derive 86% NEAshaving io251, so we adopt the same percentage for our statisticalanalysis. The area of the celestial sphere between ecliptic latitudes�251 and þ251 represents 17400 square degrees, namely about42% from the total area of a sphere. Extrapolating our number of 11objects observed in 24 square degrees within the ecliptic zone�251oβo þ 251, we derive about 8000 unknown best NEOcandidates having io251. Extrapolating in turn this number toall inclinations, we conclude that there are in total some 9300unknown NEO candidates observable with a 2 m survey. Addingthis number to the currently known almost 10000 NEAs (discov-ered by 1–2 m surveys), we predict about 19300 NEAs observableby 2 m surveys. This number is very close to Mainzer and et al.(2012) who used NEOWISE data to predict a total of 20;50074200NEAs larger than 100 m, which could be regarded as a limit for 2 msurveys.

5. Conclusions and future work

In this paper we report on the follow-up and recovery of 477program NEAs, PHAs and VIs observed mostly between 2010 and2012 from nine sites of the EURONEAR network which include sixprofessional 1–4 m class telescopes located in Chile, La Palma,France and Germany plus three smaller educational and publicoutreach telescopes in Romania and Germany. The addition ofthese objects to our previous work leads to a total of 739 NEAsfollowed-up by the EURONEAR network since 2006.

The reduced data presented in this paper generated 98 MPCpublications. We use the most important data to study the orbitimprovement of 111 NEAs, PHAs and VIs, among which 29 objectswere recovered at a new opposition and the other 82 werefollowed-up soon after discovery. Although recovery of NEAs afew years after their last observation appears to be important,follow-up of newly discovered objects soon after discovery seemsto be more valuable for the recovery and orbital improvement offainter and highly uncertain objects which could remain invisibleto existing surveys.

We characterize each site based on the astrometric residualplot for all observed NEAs, finding WHT-ACAM, INT-WFC andBlanco-MOSAIC II (close to center only) the best instruments (RMSof the O�C 0:44–0:46″), followed by Bonn and Galati (0:56–0:57″),particularly good small educational facilities. Using our publisheddata of known MBAs, we compare the astrometry across the largefield of the INT-WFC and Blanco-MOSAIC II and also the astro-metric improvement over the whole field of the INT-WFC afterimage correction of its quite distorted prime focus field, reachingRMS 0:41″, more than two times better than 0:97″, our past resultswithout THELI correction.

During two runs plus a few other discretionary hours (equiva-lent to five clear nights in total) we used the large field INT-WFCand Blanco-MOSAIC II facilities to recover and follow-up 47 highlyuncertain NEAs and also to carry out an opposition mini-surveywith the INT (10 square degrees) focused on MBAs. In 115 fieldsobserved with these two 2–4 m large field facilities covering atotal of 34 sky square degrees we carefully measured and reportedto the MPC, in addition to the 47 program NEAs, all identifiedmoving objects, comprising 868 known MBAs, 986 unknown

O. Vaduvescu et al. / Planetary and Space Science 85 (2013) 299–311310

objects—mostly MBAs, and 104 newly discovered MBAs. We usethese data to derive new MBA and NEA observability statistics for2–4 m surveys, continuing our previous work based on dataobtained using 1–2 m large field facilities. The INT or any similar2 m-class telescope can observe NEAs as faint as R∼22, being mostefficient in discovery of MBAs around R∼21, while Blanco or asimilar 4 m class facility can detect NEAs up to R∼24, being mostefficient to discover new objects around R∼22:5. Based on our INTdataset, a total of 107 MBAs per square degree can be detected atopposition up to R∼22:0 in a 2 m survey (this density surpassingour 2 m past results but remaining very close to other resultsbased on surveys using larger telescopes). The two best fieldsobserved with Blanco in the ecliptic in good weather conditionsresult in 135 MBAs per square degree to R∼23.

The ratio of unknown to known MBAs observable by a 2 m or4 m class survey is 1.4 based on our INT data (very consistent withour past work) and 2.7 based on a few fields observed in goodweather with Blanco, which could discover twice as manyunknown MBAs with respect to known MBAs than the INT.

About 104 new MBAs were recovered in two or three nightsduring our opposition mini-survey using the INT, so we expectthem to be credited in the future as EURONEAR MBA discoveries.We studied our 986 one-night unknown objects observed withBlanco and the INT using three independent NEO search criteria,finding 60 NEO candidates satisfying at least one criterion and 18best NEO candidates satisfying at least two. Using our total skycoverage (24 square degrees with INT and 10 square degrees withBlanco) and counting the best NEO candidates, we derived anaverage of 0.5 NEO candidates per square degree observable in a2 m survey (in very good agreement with our past results) and atleast 0.7 NEO candidates per square degree based on our oneBlanco night partially affected by clouds. The ratio of unknownobjects over best NEO candidates observed with INT and Blanco is60, in very good agreement with the ratio of 66 obtained whenusing the entire known published asteroid and NEA populations.Using the11 best NEO candidates observed with the INT in 24 squaredegrees and a simple two step orbital model, we assess the totalnumber of NEAs detectable by a2 m survey to19300 objects, in veryclose agreement with a recent work.

We conclude by listing two future projects aiming to continuethe study of the NEA distribution at the faint end, using data fromexisting large field 4–8 m class telescopes. The first facility is thenew DECam camera installed on the Blanco telescope. Second isthe SuprimeCam camera on Subaru (based on archival data relatedto another EURONEAR data-mining project) and also the newSubaru Hyper-SuprimeCam camera, which could yield extensivestatistics allowing existing NEO models to be checked and theformation and evolution of the entire NEA population to bestudied. These facilities could be scanned using automated detec-tion software, then comparing results with human search based onour experience and other independent studies.

Acknowledgments

Special thanks are due to the science committees and institu-tions awarding time to our EURONEAR proposals: the ChileanNational Time Committee (for the Blanco observing run 0646, 3–4June 2011) and the Spanish Time Allocation Committee (for theINT observing run C6, 25–28 February 2012). Special thanks aredue to the Isaac Newton Group (ING), the U.K. Science andTechnology Facilities Council (STFC), the Spanish “Ministerio deCiencia e Innovación” (MICINN) and the “Instituto Astrofisico deCanarias” (IAC) for supporting the project AYA2008-06202-C03-02thanks to six visiting students who received funding for their INTobserving run. M. Karami thanks the Iranian National Observatory

project for the INT training opportunity and the INT observingnights. To correct the optical field distortions of the INT-WFCimages we used THELI software; thanks are due to M. Schirmer forhis assistance in using it. Acknowledgments are due to Bill Gray forproviding, developing and allowing free usage of FIND_ORB to theentire amateur-professional community. We thank the MinorPlanet Center, especially T. Spahr and G. Williams who revisedour MPC reports. Thanks are also due to the two referees whoprovided feedback important to improve our paper. This researchhas made intensive use of the Astrometrica software developed byHerbert Raab, very simple to install and use by students andamateur astronomers. We also used the image viewer SAOImageDS9, developed by Smithsonian Astrophysical Observatory andalso IRAF, distributed by the National Optical Astronomy Observa-tories, operated by the Association of Universities for Research inAstronomy, Inc. under cooperative agreement with the NationalScience Foundation.

Appendix A. Supplementary data

Supplementary data associated with this article can be found inthe online version at http://dx.doi.org/10.1016/j.pss.2013.06.026.

References

Apitzsch, R., et al., 2011. MPEC 2011-E19.Arlot, J.E., et al., 2010a. MPC 69732, 12.Arlot, J.E., et al., 2010b. MPC 72456, 4.Bacci, P., et al., 2012. MPEC 2012-B17.Balam, D.D., et al., 2011. MPC 75940, 4.Birlan, M., et al., 2010a. Astronomy & Astrophysics 511, 40.Birlan, M., et al., 2010b. MPC 70198, 9.Birlan, M., et al., 2010c. MPEC 2010-W11.Birlan, M., et al., 2010d. MPEC 2010-W12.Birlan, M., et al., 2010e. MPEC 2010-W13.Birlan, M., et al., 2011a. MPEC 2011-E11.Birlan, M., et al., 2011b. MPEC 2011-E12.Birlan, M., et al., 2011c. MPEC 2011-E13.Birlan, M., et al., 2011d. MPEC 2011-W29.Birlan, M., et al., 2011e. MPEC 2011-W33.Borngen, F., Stecklum, B., 2012a. MPC 79140, 1.Borngen, F., Stecklum, B., 2012b. MPC 79473, 5.Bottke, W.F., et al., 2006. The Annual Review of Earth and Planetary Sciences 34,

157.Bowell, E., Muinonen, K., 1994. Earth-crossing asteroids and comets: groundbased

search strategies. In: Gehrels, S. (Ed.), Hazards Due to Comets and Asteroids.University Arizona Press, p. 149.

Bressi, T.H., et al., 2012. MPEC 2012-D102.Buie, M.W., et al., 2011a. MPC 74496, 6.Buie, M.W., et al., 2011b. MPC 75198, 5.Buie, M.W., et al., 2011c. MPC 75623, 6.Buie, M.W., et al., 2011d. MPC 75939, 2.Buie, M.W., et al., 2011e. MPC 76865, 5.Buie, M.W., et al., 2012. MPC 79528, 6.Buzzi, L., et al., 2011a. MPEC 2011-W12.Buzzi, L., et al., 2011b. MPEC 2011-W22.Buzzi, L., et al., 2011c. MPEC 2011-W27.Buzzi, L., et al., 2011d. MPEC 2011-W28.Buzzi, L., et al., 2011e. MPEC 2011-W44.Buzzi, L., et al., 2011f. MPEC 2011-W45.Buzzi, L., et al., 2011g. MPEC 2011-W52.Cavadore, C., et al., 2011a. MPC 74036, 3.Cavadore, C., et al., 2011b, MPC 77173, 6.Christensen, E.J., et al., 2012. MPEC 2012-J34.Eglitis, I., et al., 2012. MPEC 2012-H42.Elst, E.W., et al., 2011a. MPC 76441, 4.Elst, E.W., et al., 2011b. MPC 77265, 3.Erben, T., et al., 2005. Astronomische Nachrichten 326, 432. (THELI software).EURONEAR, 2012a. The European Near Earth Asteroids Research ⟨http://euronear.

imcce.fr⟩.EURONEAR, 2012b. Observing Tools: Planning ⟨http://euronear.imcce.fr/tiki-index.

php?page=Planning⟩.Farinella, P., et al., 1993. Icarus 101, 174.Fitzsimmons, A., et al., 2011a. MPC 76443, 2.Fitzsimmons, A., et al., 2011b. MPC 74148, 2.Fitzsimmons, A., et al., 2011c. MPC 74893, 1.

O. Vaduvescu et al. / Planetary and Space Science 85 (2013) 299–311 311

Fitzsimmons, A., et al., 2011d. MPC 75207, 2.Fitzsimmons, A., et al., 2011e. MPC 77266, 1.Fitzsimmons, A., et al., 2012a. MPC 78047, 5.Fitzsimmons, A., et al., 2012b. MPC 78437, 11.Fitzsimmons, A., et al., 2012c. MPC 78894, 9.Fitzsimmons, A., et al., 2012d. MPC 79221, 1.Fitzsimmons, A., et al., 2012e. MPC 79530, 1.Gajdos, S., et al., 2012, MPEC 2012-E19.Gray, B., 2012, FIND_ORB ⟨http://www.projectpluto.com/find_orb.htm⟩.Holman, M., et al., 2011. MPC 75625, 2.Jaeger, M., et al., 2012. MPEC 2012-H90.Lang, D., et al., 2010. Astronomical Journal 139, 1782L.Lehmann, G., et al., 2011. MPEC 2011-E14.Mainzer, A., et al., 2012. The Astrophysical Journal 752, 110.McMillan, R., et al., 2011, MPEC 2011-W25.Milani, A., Gronchi, G.F., 2009. Theory of Orbit Determination.Morbidelli, A., et al., 2002. Origin and evolution of near-earth objects. In: Bottke,

et al. (Eds.), Asteroids III. University Arizona Press, p. 409.Minor Planet Center (MPC), 2012a. ⟨http://www.minorplanetcenter.net⟩.MPC, 2012b. NEO Rating ⟨http://www.minorplanetcenter.net/iau/NEO/PossNEO.

html⟩.Muinonen, K., Bowell, E., Wasserman, L.H., 1994. Planetary and Space Science 42

(4), 307.NASA JPL, 2012. Sentry Risk Table ⟨http://neo.jpl.nasa.gov/risk⟩.Nishiyama, K., et al., 2011. MPEC 2012-J01.Near Earth Objects—Dynamic Site (NEODyS), 2012 ⟨http://newton.dm.unipi.it/

neodys⟩.Raab, H., 2012. Astrometrica Software ⟨http://www.astrometrica.at⟩.Sonka, A., 2008. MPC 64104, 11.Sonka, A., 2009a. MPC 65048, 3.Sonka, A., 2009b. MPC 65638, 6.Sonka, A., 2009c. MPC 65928, 13.Sonka, A., 2009d. MPC 67139, 5.Sonka, A., 2009e. MPC 67404, 7.Sonka, A., 2009f. MPC 67677, 8.Sonka, A., 2010a. MPC 69817, 4.Sonka, A., 2010b. MPC 71094, 4.Sonka, A., 2010c. MPC 72053, 3.Sonka, A., 2011a. MPC 75208, 6.

Sonka, A., 2011b. MPC 75447, 3.Sonka, A., 2011c. MPC 77270, 7.Sonka, A., 2012a. MPC 77702, 7.Sonka, A., 2012b. MPC 78050, 8.Sonka, A., 2012c. MPC 79998, 13.Spaceguard System, 2012. Priority List ⟨http://spaceguard.iaps.inaf.it/SSystem/SSys

tem.html⟩.Stecklum, B., 2012a. MPC 79746, 2.Stecklum, B., 2012b. MPC 79971, 5.Stecklum, B., et al., 2012a. MPEC 2012-K07.Stecklum, B., et al., 2012b. MPEC 2012-G45.Tercu, O., Dumitriu, A., 2011a. MPC 75947, 2.Tercu, O., Dumitriu, A., 2011b. MPC 76451, 8.Tercu, O., Dumitriu, A., 2011c. MPC 76874, 7.Tercu, O., Dumitriu, A., 2011d. MPC 77273, 6.Tercu, O., Dumitriu, A., 2012a. MPC 77705, 1.Tercu, O., Dumitriu, A., 2012b. MPC 78053, 2.Tercu, O., Dumitriu, A., 2012c. MPC 78445, 7.Tercu, O., Dumitriu, A., 2012d. MPC 78902, 10.Tercu, O., Dumitriu, A., 2012e. MPC 79225, 1.Tercu, O., Dumitriu, A., 2012f. MPC 79533, 5.Tercu, O., Dumitriu, A., 2012g. MPC 80000, 11.Tudorica, A., 2011. MPC 74509, 1.Tudorica, A., 2012a. MPC 79224, 10.Tudorica, A., 2012b. MPC 79789, 7.Tudorica, A., 2012c. MPC 80000, 10.Tudorica, A., 2012d. MPC 77273, 5.Tudorica, A., 2012e. MPC 79533, 4.Tudorica, A., Badescu, T., 2011. MPC 76874, 4.Tudorica, A., Badescu, T., 2012. MPC 78902, 8.Tudorica, A., Toma, R., 2011. MPC 76451, 6.Vaduvescu, O., et al., 2008. Planetary and Space Science 56 (14), 1913.Vaduvescu, O., et al., 2011. Planetary and Space Science 59 (13), 1632.Vaduvescu, O., et al., 2012a. MPC 79787, 2.Vaduvescu, O., et al., 2012b. MPEC 2012-D82.Williams, G., private communication.Yoshida, F., et al., 2003. Publications of the Astronomical Society of Japan 55, 701.Yoshida, F., Nakamura, T., 2007. Planetary and Space Science 55, 1113.